SICE: an improved missing data imputation technique
Abstract In data analytics, missing data is a factor that degrades performance. Incorrect imputation of missing values could lead to a wrong prediction. In this era of big data, when a massive volume of data is generated in every second, and utilization of these data is a major concern to the stakeh...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
SpringerOpen
2020-06-01
|
Series: | Journal of Big Data |
Subjects: | |
Online Access: | http://link.springer.com/article/10.1186/s40537-020-00313-w |
id |
doaj-6422474a094445f98f45cec877277047 |
---|---|
record_format |
Article |
spelling |
doaj-6422474a094445f98f45cec8772770472020-11-25T03:20:48ZengSpringerOpenJournal of Big Data2196-11152020-06-017112110.1186/s40537-020-00313-wSICE: an improved missing data imputation techniqueShahidul Islam Khan0Abu Sayed Md Latiful Hoque1Department of CSE, Bangladesh University of Engineering and TechnologyDepartment of CSE, Bangladesh University of Engineering and TechnologyAbstract In data analytics, missing data is a factor that degrades performance. Incorrect imputation of missing values could lead to a wrong prediction. In this era of big data, when a massive volume of data is generated in every second, and utilization of these data is a major concern to the stakeholders, efficiently handling missing values becomes more important. In this paper, we have proposed a new technique for missing data imputation, which is a hybrid approach of single and multiple imputation techniques. We have proposed an extension of popular Multivariate Imputation by Chained Equation (MICE) algorithm in two variations to impute categorical and numeric data. We have also implemented twelve existing algorithms to impute binary, ordinal, and numeric missing values. We have collected sixty-five thousand real health records from different hospitals and diagnostic centers of Bangladesh, maintaining the privacy of data. We have also collected three public datasets from the UCI Machine Learning Repository, ETH Zurich, and Kaggle. We have compared the performance of our proposed algorithms with existing algorithms using these datasets. Experimental results show that our proposed algorithm achieves 20% higher F-measure for binary data imputation and 11% less error for numeric data imputations than its competitors with similar execution time.http://link.springer.com/article/10.1186/s40537-020-00313-wMissing Data ImputationSingle ImputationMultiple ImputationMICEData Analytics |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Shahidul Islam Khan Abu Sayed Md Latiful Hoque |
spellingShingle |
Shahidul Islam Khan Abu Sayed Md Latiful Hoque SICE: an improved missing data imputation technique Journal of Big Data Missing Data Imputation Single Imputation Multiple Imputation MICE Data Analytics |
author_facet |
Shahidul Islam Khan Abu Sayed Md Latiful Hoque |
author_sort |
Shahidul Islam Khan |
title |
SICE: an improved missing data imputation technique |
title_short |
SICE: an improved missing data imputation technique |
title_full |
SICE: an improved missing data imputation technique |
title_fullStr |
SICE: an improved missing data imputation technique |
title_full_unstemmed |
SICE: an improved missing data imputation technique |
title_sort |
sice: an improved missing data imputation technique |
publisher |
SpringerOpen |
series |
Journal of Big Data |
issn |
2196-1115 |
publishDate |
2020-06-01 |
description |
Abstract In data analytics, missing data is a factor that degrades performance. Incorrect imputation of missing values could lead to a wrong prediction. In this era of big data, when a massive volume of data is generated in every second, and utilization of these data is a major concern to the stakeholders, efficiently handling missing values becomes more important. In this paper, we have proposed a new technique for missing data imputation, which is a hybrid approach of single and multiple imputation techniques. We have proposed an extension of popular Multivariate Imputation by Chained Equation (MICE) algorithm in two variations to impute categorical and numeric data. We have also implemented twelve existing algorithms to impute binary, ordinal, and numeric missing values. We have collected sixty-five thousand real health records from different hospitals and diagnostic centers of Bangladesh, maintaining the privacy of data. We have also collected three public datasets from the UCI Machine Learning Repository, ETH Zurich, and Kaggle. We have compared the performance of our proposed algorithms with existing algorithms using these datasets. Experimental results show that our proposed algorithm achieves 20% higher F-measure for binary data imputation and 11% less error for numeric data imputations than its competitors with similar execution time. |
topic |
Missing Data Imputation Single Imputation Multiple Imputation MICE Data Analytics |
url |
http://link.springer.com/article/10.1186/s40537-020-00313-w |
work_keys_str_mv |
AT shahidulislamkhan siceanimprovedmissingdataimputationtechnique AT abusayedmdlatifulhoque siceanimprovedmissingdataimputationtechnique |
_version_ |
1724616589028884480 |